Capstone project : ISB <> Peakhealth

Demo link 📺

Click to view the demo

How to contribute to the repo 🤔

• Create a separate branch for your usecase
• Rebase with main branch
• Raise PR (DO NOT commit to main)

Info how to make best use of the workflow ⏯

• Clone the repo
• Create the env using the above commands
• Install the package
• All the tests goes into the tests directory
• Any test experiments eg- Data wraggling/inital exploration notebooks goes under experiments directory
• All modelling and related details into analysis_framework, create subdirectories as required
• Under forecasting_framework there are three submodules utils, data and pipelines
  • utils reusable code components
  • data raw dataset are stored
  • pipeline dagster pipeline and related computation output

Flowchart for repo structure

More on Git/version control usage

Material link

Repository setup

• Set up authentication using OAuth, SSH or push token

# for the first time
git clone <repo>

# change working directory to root of repo cloned
cd <repo location>

Installing env setup using miniconda 💁🏽‍♀️

• Instructions

Commands to install the packages via conda/ env activation ✍️

conda env create --file environment.yml

conda activate capstone

conda list

conda info

Installing the analysis_framework package in local ⬇️

pip install -e .

Update the capstone environment (post making changes to environment file)

conda env update --file environment.yml --prune

To switch between different environment

conda deactivate

About pre-commit-hooks and activating 🔌

Just like the name suggests, precommit-hooks are designed to format the code based on PEP standards before committing. More details 🗒

pip install pre-commit
pre-commit install

"""---commit your changes---
- check for errors, hooks format the code by deafult
- add the files
- commit it again
- push the changes 
"""

Data pipeline 🛠

• Built using Dagster, an open source orchestration tool

Commands to trigger the pipeline

• Command to run the pipeline from root capstone_isb

python analysis_framework/pipeline/process.py

• Command to view the dagit UI

dagit -f analysis_framework/pipeline/process.py

Dashboard

• Tools used - Panel, Holoviews
• Main files app.py (panel serve requirements and layout components) and base.py (contains all the logic)
• Command to trigger the dashboard (subject to change) - python analysis_framework/dashboard/app.py